Chien Van Trinh

Edge-preserving block compressive sensing with projected landweber

Compressive Sensing (CS) is an emerging new sampling technique which helps to break through the Nyquist sampling frequency for sparse signals. This paper addresses improving one of its recovery algorithms known as the Block Compressive Sensing with Smooth Projected Landweber (BCS-SPL). For reducing the blocking artifacts in BCS-SPL, the Wiener filter has been implemented as a classic way to smooth image at the beginning of each iteration, but it is quite sensitive to image edges and blurs the image. In this paper, we propose a modified method which separates image signal into its low and high frequency components, and then independently processes each of the two components. Subsequently, a smoothness enhancing operation is implemented to improve reduction of high frequency oscillatory artifacts after hard thresholding. Experimental results show that the proposed method improves reconstructed image quality by more than 3dB compared to the conventional BCS-SPL.

Measurement Coding for Compressive Sensing of Color Images

From the perspective of reducing the sampling cost of color images at high resolution, block-based compressive sensing (CS) has attracted considerable attention as a promising alternative to conventional Nyquist/Shannon sampling. On the other hand, for storing/transmitting applications, CS requires a very efficient way of representing the measurement data in terms of data volume. This paper addresses this problem by developing a measurement-coding method with the proposed customized Huffman coding. In addition, by noting the difference in visual importance between the luma and chroma channels, this paper proposes measurement coding in YCbCr space rather than in conventional RGB color space for better rate allocation. Furthermore, as the proper use of the image property in pursuing smoothness improves the CS recovery, this paper proposes the integration of a low pass filter to the CS recovery of color images, which is the block-based l 20 -norm minimization. The proposed coding scheme shows considerable gain compared to conventional measurement coding.

Block-based compressive sensing of video using local sparsifying transform

Block-based compressive sensing is attractive for sensing natural images and video because it makes large-sized image/video tractable. However, its reconstruction performance is yet to be improved much. This paper proposes a new block-based compressive video sensing recovery scheme which can reconstruct video sequences with high quality. It generates initial key frames by incorporating the augmented Lagrangian total variation with a nonlocal means filter which is well known for being good at preserving edges and reducing noise. Additionally, local principal component analysis (PCA) transform is employed to enhance the detailed information. The non-key frames are initially predicted by their measurements and reconstructed key frames. Furthermore, regularization with PCA transform-aided side information iteratively seeks better reconstructed solution. Simulation results manifest effectiveness of the proposed scheme.

Automatic number plate recognition on electronic toll collection systems for Vietnamese conditions

Due to unlimited increase of cars and transportation systems, a real time embedded system called Automatic Number Plate Recognition (ANPR) is very important for humans to detect and manage. This paper presents results of developing and deploying an ANPR applied to electronic tolling collection (ETC) systems in Vietnam with some special issues. Our model is designed and investigated by using a VIVOTEK IP8361 camera to capture an image. After that, the image is transmitted to an industrial computer to process. In detail, the image is processed first to reduce noise and artifacts by a low-pass filter before our software detects plate candidates. The characters in the candidates are then extracted by an optical character recognition utilizing neural network. We also employ Microsoft visual C sharp integrated development environment to build graphical user interface. Experimental results manifest the high accuracy of our method achieving approximately 85.00% and the processing time of only about 20--30ms.

A Skip-mode Coding for Distributed Compressive Video Sensing

분산 압축 비디오 센싱 (DCVS) 기술은 압축센싱 및 분산 비디오 부호화 기술의 결합을 통해 저 비용의 샘플링을 실현하는 새로운 패러다임이다. 본 논문에서는 프레임 간 높은 시간 상관성을 활용한 DCVS에서의 스킵모드 부호화 방법을 제안한다. 제안하는 방법은 일정조건을 만족하는 비 키-프레임에 대한 측정값을 복호화기에 전송하지 않아도 시간적 보간법을 통해 해당 비 키-프레임의 복원이 가능하도록 하여 율-왜곡 측면에서 좋은 압축 성능을 보장한다. 이와 더불어, 더 나은 시간적 보간을 위하여 계층적 구조를 사용하는 방법을 제안한다. 실험 결과, 제안하는 스킵모드 부호화 방법은 약간의 PSNR 감소에 비해 매우 높은 측정율 절약이 되는 것을 확인하였다. 또한, 제안하는 방법을 높은 시간 연관성을 갖는 비디오 영상에 적용할 경우, 복호화기의 연산 복잡도가 평균 43.75% 감소하는 것을 확인하였다.

Compressive Sensing recovery with improved hybrid filter

Compressive Sensing (CS) is a novel sampling framework which is more efficient than the Nyquist sampling for sparse signals. A major challenge in CS is its quality improvement of recovered signal when noise exists. To reduce noise in the recovered images, filters are usually employed. This paper focuses on improving the quality of CS recoveries by applying a hybrid filter which pursues smoothness and preserves edge at the same time. Considering desirability of the block-based recovery in practical usages, the proposed hybrid filter is investigated not only for the frame-based recovery but also for the block-based recovery. Experimental results demonstrate that the proposed hybrid filter attains much better performance in CS recovery than the conventional ones in term of both subjective and objective qualities.

Rate Allocation for Block-based Compressive Sensing

희소성이 높은 신호를 압축센싱을 할 경우 기존의 Nyquist/Shannon 이론을 바탕으로 하는 샘플링 방법 보다 낮은 측정율 만으로도 신호의 복원이 가능하기 때문에 이를 활용한 많은 응용 연구가 이루어지고 있다. 영상신호의 경우 특히 블록기반 압축센싱 기법이 주로 고려되고 있는데, 대부분의 경우 측정 영역에서의 공간적 유사도가 동일하다는 가정 하에, 각 블록에 동일한 측정율을 할당하여 왔다. 이를 개선하기 위해, 본 논문에서는 프레임 내의 각 블록에 대하여 경계선 정보를 구하고, 각각의 특성에 따르는 적응적 샘플링율 기법을 제안한다. 제안하는 방법은 측정영역에서의 블록 간 유사도를 구해서 경계선 정보를 많이 포함하는 블록일수록 많은 측정율을 할당한다. 실험 결과, 자연영상에 대해 제안하는 적응적 율 할당 기법은 고정 측정율을 사용한 기존 방법에 비해 객관적 (최대 3.29 dB 향상) 및 주관적 화질이 뛰어나다는 것을 보여준다.

Implementation of a MIMO-OFDM system based on the TI C64x+ DSP

In this paper, we present an implementation of a multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) system based on the multi-core Texas Instrument (TI) C64x+ digital signal processor (DSP). The system is implemented by employing real time data exchange (RTDX) and serial rapid input/output (SRIO) techniques for communication interfaces between personal computer (PC) and DSP. Received signals obtained by the DSP processing are visualized to verify the implemented system. Furthermore, this study is concerned with bit error rate (BER) performance and the computational complexity of system written in C/C++.